Taxi High-income Regions Mining And Influencing Factors Analysis Based On GPS Data

With the popularity of Internet ride-hailing and increasing urban traffic congestion,taxi strikes have occurred from time to time.As an important supplement to urban public transportation,taxis account for a relatively high proportion of urban passenger transportation,and their income level is the focus of the attention of taxi drivers and regulatory agencies.Taxi trajectory data can reflect the time and space rules of taxi passengers to a certain extent.At present,most research results study how to increase the number of taxi passengers.There is a lack of research on how to improve taxi passenger revenue.This article analyzes taxi travel characteristics and Comparing the similarities and differences in the spatial distribution of passenger hotspot areas and high-income passenger-carrying areas can be used as a reference for increasing the income of taxi drivers and making policies for urban planning and management departments.Firstly,this paper summarizes the current situation for the development of this research by summarizing and analyzing the previous research literature.Based on the taxi GPS track data of Xi ’an in October,and in order to ensure the rationality and applicability of the taxi operation data,the data are preprocessed and the effective data are output.Secondly,the spatio-temporal distribution characteristics of taxi travel are analyzed.From the perspective of time,this paper analyzes the time-varying characteristics of taxi efficiency operating index,order income,and passenger-carrying distance,and points out the differences of each index in holidays,working days,and non-working days.From the spatial perspective,the spatial distribution characteristics of passenger carrying events in the study area were analyzed based on travel demand.Then,the grey target decision-making model is established to determine the research period of this paper through the operational efficiency index.By constructing Tyson’s polygon,the research area of this paper is divided into 818 regular hexagons,and the target area of this paper is redefined according to the research content of this paper.A focal statistics algorithm was used to determine the recommended target areas at different dates and different times in this paper,and the similarities and differences between the Spatio-temporal distribution characteristics of passenger carrying hot spots and target areas were analyzed.The analysis results from time and space latitudes show that the spatial distribution of the target region varies greatly in holidays,working days,and non-working days,and the variation of period time produces great fluctuations.Most of the target regions are spatiotemporal non-stationary.Passenger hotspots have not always been high-income regions.Finally,the POI data and regional order data attributes are used as independent variables to be included in the alternative influencing factors of the passenger area.The multicollinearity test of the independent variables shows that there is a certain degree of multicollinearity between the independent variables.The principal component analysis method is used to determine the influencing factors introduced into the regression model to be attributed to order attributes,life service facilities,leisure and entertainment facilities,and public service facilities.Transportation infrastructure,landscape facilities.The results show that the above influencing factors have a significant correlation with the income level of the passenger-carrying area;the multivariate ordered Logistic model is more effective than the linear regression model.Next,construct the Fisher discriminant forecast model to predict the regional income level,compare the forecast result with the actual result,and the result shows that the accuracy of the forecast model reaches 90%,which verifies the validity of the Fisher discriminant forecast model.The above research provides a basic reference for drivers to search for passengers to increase revenue,improve the operating efficiency of the taxi industry,and build urban transportation infrastructure.